ECML PKDD 2018 Workshops MIDAS 2018 and PAP 2018, Dublin, Ireland, September 10-14, 2018, Proceedings

This book constitutes revised selected papers from two workshops held at the 18th European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2018, in Dublin, Ireland, in September 2018, namely:MIDAS 2018 - Third Workshop on Mining Data for Financial Applications andPAP 2...

Full description

Saved in:
Bibliographic Details
Main Authors Alzate, Carlos, Monreale, Anna, Bioglio, Livio, Bitetta, Valerio, Bordino, Ilaria, Caldarelli, Guido, Ferretti, Andrea, Guidotti, Riccardo, Gullo, Francesco, Pascolutti, Stefano
Format eBook
LanguageEnglish
Published Netherlands Springer Nature 2019
Springer International Publishing AG
Springer
Edition1
SeriesLecture Notes in Computer Science
Subjects
Data mining
Special computer methods
Online AccessGet full text

Cover

Table of Contents:
  • PAP 2018: The 2nd International Workshop on Personal Analytics and Privacy -- Workshop Description -- Part2 -- PAP Chairs -- Program Committee -- A Differential Privacy Workflow for Inference of Parameters in the Rasch Model -- 1 Introduction -- 2 Preliminaries -- 3 Methods -- 4 Experiments -- 4.1 Results -- 5 Conclusion -- References -- Privacy Preserving Client/Vertical-Servers Classification -- 1 Introduction -- 2 Background -- 3 Client/Vertical-Servers Random Forest -- 3.1 Problem Setting -- 3.2 Client/Vertical-Servers Set Intersection Protocol -- 3.3 Building a Privacy-Preserved Random Forest -- 3.4 Client/Vertical-Servers Random Forest Classification -- 3.5 Security Analysis -- 4 Experimental Results and Complexity -- 4.1 Experiment Settings and Datasets -- 4.2 Complexity Comparison -- 4.3 Performance Analysis -- 5 Conclusion -- References -- Privacy Risk for Individual Basket Patterns -- 1 Introduction -- 2 Related Work -- 3 Retail Data -- 4 Privacy Risk Assessment Methodology -- 5 Experiments -- 5.1 Patterns Against Patterns -- 5.2 Patterns Against Baskets -- 6 Conclusion -- References -- Exploring Students Eating Habits Through Individual Profiling and Clustering Analysis -- 1 Introduction -- 2 Related Work -- 3 Food Dataset -- 4 Food Consumption Analytics -- 4.1 Data Preprocessing -- 4.2 Clustering Analysis -- 5 Temporal Food Habits Analytics -- 5.1 Clustering Analysis -- 6 Conclusion -- A Appendix -- References -- Author Index
  • Intro -- Preface -- Contents -- MIDAS 2018: The 3rd Workshop on MIning DAta for Financial ApplicationS -- Workshop Description -- Part1 -- Program Chairs -- Program Committee -- A Multivariate and Multi-step Ahead Machine Learning Approach to Traditional and Cryptocurrencies Volatility Forecasting -- 1 Introduction -- 2 Dynamic Factor Machine Learner -- 3 Methodology -- 3.1 Multivariate Forecasting Methods -- 3.2 Datasets Description -- 3.3 Volatility Proxies -- 4 Experimental Results -- 4.1 Results Discussion -- 5 Conclusion and Future Work -- References -- Calibrating the Mean-Reversion Parameter in the Hull-White Model Using Neural Networks -- 1 Introduction -- 2 Background and Related Work -- 2.1 Hull-White Interest Rate Model -- 2.2 Neural Networks -- 2.3 Related Work -- 3 Mean Reversion in the Hull-White Model -- 4 Methodology -- 4.1 Can Theta Be Replaced? -- 4.2 Mapping -- 4.3 Evaluation and Datasets -- 5 Results -- 6 Conclusion -- References -- Deep Factor Model -- 1 Introduction -- 2 Related Works -- 3 Methodology - Deep Factor Model -- 3.1 Deep Learning -- 3.2 Layer-Wise Relevance Propagation -- 3.3 Deep Factor Model -- 4 Experiment on Japanese Stock Markets -- 4.1 Data -- 4.2 Model -- 4.3 Results -- 4.4 Interpretation -- 5 Conclusion -- References -- A Comparison of Neural Network Methods for Accurate Sentiment Analysis of Stock Market Tweets -- 1 Introduction -- 2 Data -- 2.1 Labeling Using Amazon Mechanical Turk -- 3 Method and Models -- 3.1 Preprocessing -- 3.2 Word Embeddings -- 3.3 Baseline Model -- 4 Neural Network Models -- 4.1 Convolutional Neural Networks -- 4.2 Experimenting with Recurrent Neural Networks -- 5 Results -- 6 Comparing the Sentiments with Stock Market Returns -- 6.1 Granger Causality Models -- 6.2 Our Granger Causality Model -- 6.3 Three Year Comparison of Social Media Sentiment Analysis and Stock Market Returns
  • 7 Conclusion -- References -- A Progressive Resampling Algorithm for Finding Very Sparse Investment Portfolios -- 1 Introduction -- 2 Background and Problem Statement -- 2.1 The Markowitz Model and Its L1 Regularised Variant -- 2.2 Naive k-portfolios -- 3 A Progressive Resampling Algorithm -- 3.1 Basic Idea -- 3.2 Why Should Progressive Resampling Work? -- 4 Experiments -- 4.1 Datasets and Baseline Methods -- 4.2 Experiment 1: Are There Any Reasonable Naive k-Portfolios? -- 4.3 Experiment 2: Out-of-Sample Performance of Naive k-Portfolios -- 5 Discussion, Limitations, and Conclusions -- 5.1 Related Approaches -- 5.2 Generalising the Naive k-portfolio Problem -- 5.3 Limitations -- 5.4 Conclusion and Future Work -- References -- ICIE 1.0: A Novel Tool for Interactive Contextual Interaction Explanations -- 1 Introduction -- 2 Related Work -- 3 Preliminaries -- 4 The ICIE Method -- 4.1 Calculation of Contextual SHAP Values -- 4.2 UI for Context Exploration -- 5 Use Cases -- 5.1 Adult -- 5.2 Credit-Card-Default -- 5.3 German-Credit -- 5.4 Mushroom -- 5.5 Tic-Tac-Toe -- 5.6 Wisconsin -- 5.7 Guiding the Manual Search for Explanations -- 6 Discussion -- References -- Testing for Self-excitation in Financial Events: A Bayesian Approach -- 1 Introduction -- 2 Model -- 2.1 Hawkes Process -- 2.2 Bayesian Model Comparison -- 2.3 Proposed Method -- 3 Experiments -- 4 Conclusion -- References -- A Web Crawling Environment to Support Financial Strategies and Trend Correlation -- 1 Introduction -- 2 Web Crawling and Web Data Analysis in ENEAGRID -- 2.1 Web Crawling Tool: BUbiNG -- 2.2 Virtual Laboratory and Web Application -- 2.3 Tests and Experimental Results -- 3 Proposal of Current Development -- 3.1 Thematic Web Crawling -- 3.2 Web Crawling for Financial Strategies -- 4 Conclusions -- References